ui-tars-1.5-7b-idm-full-sft-diff

This model is a fine-tuned version of ByteDance-Seed/UI-TARS-1.5-7B on the ui_tars_sft_data_diff dataset. It achieves the following results on the evaluation set:

• Loss: 0.0179

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 2
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 16
• total_eval_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 4.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.0484	1.0	281	0.0638
0.0272	2.0	562	0.0315
0.0083	3.0	843	0.0186
0.006	4.0	1124	0.0179

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.0.2
• Tokenizers 0.21.1